Kayoko Yamamoto1,*, PhD, and Shota Sashiyama2...Tokyo is the largest mega-city in the world...

In: Advances in Medicine and Biology ISBN: 978-1-53610-625-1

Editor: Leon V. Berhardt © 2017 Nova Science Publishers, Inc.

Chapter 2

AN INVESTIGATION BY BIG DATA ANALYSIS

OF THE URBAN HEAT ISLAND EFFECT

ON HUMAN HEALTH PROBLEMS

Kayoko Yamamoto1,*, PhD, and Shota Sashiyama2 1Graduate School of Informatics and Engineering,

University of Electro-Communications Tokyo, Chofu, Tokyo, Japan 2Graduate School of Information Systems,

University of Electro-Communications Tokyo, Chofu, Tokyo, Japan

ABSTRACT

The global average temperature has risen by approximately 0.7°C

over the past hundred years, and it is thought that global warming is the

main cause. However, in recent years, in addition to global warming, the

Urban Heat Island (UHI) effect has become a serious problem for major

urban areas in every country of the world. Especially in the summer, UHI

is a factor that increases damage to human health in urban areas such as

heatstroke, somnipathy, weariness, and discomfort in daily life.

Therefore, various efforts are being made to mitigate the UHI effect, and

research inquiring into causes and mitigation effects is being conducted.

For example, in recent years, the mitigating effects on higher temperature

areas of cool sea breezes which flow into inland areas are attracting

attention.

* Corresponding author: Kayoko Yamamoto. E-mail: [email protected].

The exclusive license for this PDF is limited to personal website use only. No part of this digital document may be reproduced, stored in a retrieval system or transmitted commercially in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

Kayoko Yamamoto and Shota Sashiyama 58

Against such a backdrop, focusing on the UHI which causes human

health problems, this study aims to investigate the influence on the UHI

effect of the obstruction of sea breezes by clusters of high-rise buildings

in the central Tokyo district of Japan. In the central Tokyo district,

recorded temperatures have risen by approximately 3°C in the past 100

years. To that end, this study analyzes open big data in environmental

science, using a weather simulation model and Geographic Information

Systems (GIS). In the method, two scenarios that imagine urban forms

which differ with regard to whether or not they contain high-rise

buildings are created and weather simulation is conducted, and the results

of the simulations are comparatively analyzed focusing on temperature

and wind speed. Investigation was conducted in two stages, and a region

in the central Tokyo district and a date in August 2010 were selected for

investigation.

In two stages of investigation, a rise in temperature of approximately

0.3 K and a reduction in wind speed of approximately 1 m/s were

observed in a region of 5 to 10 km square downwind of high-rise

buildings in the period 6 PM to 9 PM, and a higher temperature caused by

the obstruction of sea breeze by high-rise buildings was identified. The

fact that such a higher temperature was confirmed in the time period from

6 PM onwards, in which the temperature decreases, reveals that

obstruction of sea breeze by high-rise buildings dulls the decrease in

temperature which occurs from evening onwards, and influences

nighttime UHI formation. Therefore, the nighttime UHI causes a tropical

night which may affect human health problems.

Keywords: urban heat island (UHI), human health problems, big data

analysis, weather simulation model, geographic information systems

(GIS)

1. INTRODUCTION

According to the Japanese Ministry of the Environment (2013), the global

average temperature has risen by approximately 0.7°C over the past hundred

years, and it is thought that global warming is the main cause. However, in

recent years, in addition to global warming, the Urban Heat Island (UHI)

effect has become a serious problem for major urban areas in every country of

the world. The UHI effect is a problem of air pollution. It is a phenomenon

resulting from the fact that the temperature in central urban areas becomes

higher than that in suburban areas. Especially in the summer, through the rise

in temperature in the daytime and the increase of tropical nights, UHI is a

An Investigation by Big Data Analysis … 59

factor that increases damage to human health in urban areas such as

heatstroke, somnipathy, weariness, and discomfort in daily life (Ihara et al.,

2011). Additionally, in the summer, UHI also causes localized downpour of

short duration (so-called “guerrilla rainstorms”), and increasing energy

consumption for air conditioning (Yokoyama et al., 2014).

Therefore, various efforts are being made to mitigate the UHI effect, and

research inquiring into causes and mitigation effects is being conducted. For

example, in recent years, the mitigating effects on higher temperature areas of

cool sea breezes which flow into inland areas are attracting attention. One such

mitigation effect is achieved by the arrangement of buildings that is conscious

of ventilation paths. However, it has also been pointed out that “forests” of

high-rise buildings in coastal areas act like walls, obstructing the flow of sea

breezes into inland areas. On this account, temperature does not fall and an

UHI forms in inland areas.

In the central Tokyo district of Japan, recorded temperatures have risen by

approximately 3°C in the past 100 years. Tokyo is the largest mega-city in the

world (population 36.5 million), and has a tremendously wide range of urban

areas. Therefore, due to the fact that the temperature rise in Tokyo is the

highest among large cities all over the world, the UHI intensity as urban

warming is remarkably high in addition to the effects of global warming. A

world-famous example of an area where the forests of high-rise buildings in

coastal areas (the so-called “Tokyo Wall”) obstruct the flow of sea breezes

into inland areas is the Shiodome area of Minato City in the central Tokyo

district. Specifically, because clusters of high-rise buildings in this area block

sea breezes from Tokyo Bay, higher temperatures are not mitigated and an

UHI forms in the central Tokyo district. However, until now, the influence on

the UHI effect of the obstruction of sea breezes by clusters of high-rise

buildings has not been sufficiently understood.

Therefore, this study focuses on the UHI which causes human health

problems and the issue that although higher temperatures should be mitigated

by sea breezes flowing into inland areas, sea breeze flows are being obstructed

by forests of high-rise buildings in coastal areas. Further, this study aims to

investigate the influence on the UHI effect of the obstruction of sea breezes by

clusters of high-rise buildings in the central Tokyo district of Japan. To that

end, this study analyzes open big data in environmental science, using a

weather simulation model and Geographic Information Systems (GIS).


2. RELATED WORK

The first study on UHI to be published in Japan was Tyson et al., (1973),

and the number of studies began to increase dramatically from around the year

2000. According to Nakagawa (2011), research on the UHI effect has been

carried out in Japan since the latter half of the 1960s. In the beginning,

research mainly focused on the actual nature of the UHI effect and on

quantitatively understanding the extent of the increase in temperature. For

example, it was revealed that temperature cliffs exist in the vicinity of the

boundary of urban and suburban areas. The mainstream of research in recent

years is changing from research which places emphasis on clarifying the actual

nature of the UHI effect and the main causes of its formation to research which

aims to mitigate the UHI effect and improve amenities, by providing large-

scale green spaces, greening rooftops, improving road paving and building

materials, providing ventilation paths and so on.

However, research from the past has not lost its significance in recent

years, and there are quite a number of issues which remain unresolved

regarding clarification of the actual nature of the UHI effect and the main

causes of its formation. For example, Susca et al., (2010) evaluated the

positive effects of vegetation with a multi-scale approach: an urban and a

building scale. Focusing on anthropogenic waste heat from road traffic, which

is thought to be one of the main causes of the UHI effect, Imai et al., (2010,

2015) used a GIS to evaluate measures concerning UHIs in the 23 wards of

Tokyo in Japan. Grawe et al., (2012) focused on London in the United

Kingdom (UK), and used numerical simulation to quantitatively grasp the

influence that urbanization of land surface has on climate. Kato et al., (2015)

count on forest canopy structure to quantitatively grasp the effect of urban

forests to mitigate UHI. Uenoyama et al., (2013) showed the urban structure

containing ventilation path by simulation in the Osaka city center of Japan.

Further, as mentioned in section 1, because UHI accordingly causes the

damage to human health in urban areas, recent studies focus on the

relationship between UHI and human health problems (Tan et al., 2010, Ihara

et al., 2011, Omura et al., 2011, Kusaka et al., 2013).

Concerning studies which have focused on the mitigating effect of sea

breeze on higher temperatures, Taniguchi et al., (2008) and Junimura et al.,

(2008), and Matsuo et al., (2013) used long-term temperature and wind

observation data to understand the actual state of the mitigation of higher

temperatures by sea breezes and to understand the spatial distribution of the

area covered by the mitigation effect in Japan’s urban regions. Troy et al.,


(2009) analyzed the land use plan which mitigates the UHI phenomenon by

sea breeze in Kanagawa prefecture, Japan. Yamato et al., (2011) used

observation data on temperature and wind to examine the influence of sea

breeze on the daytime UHI effect in summer in the Greater Tokyo

Metropolitan Area of Japan. Wong et al., (2011) focused on the Kowloon

Peninsula in Hong Kong, China. From groups of buildings in the coastal area,

they selected buildings which block sea breeze as being buildings which have

a wall effect, and by conducting scenario analysis, they considered the

influence which this has on the wind environment. Takahashi et al., (2013)

analyzed the influence of the UHI phenomenon in the Tokyo Metropolis on

the nocturnal atmospheric pressure distribution and local wind system by using

observed atmospheric pressure data in the central Tokyo district and its

surroundings. Further, examples of representative preceding studies of recent

years which focused on the influence of sea breeze on the UHI effect are Ooka

et al., (2008), Nobayashi et al., (2009), Papanastasiou et al., (2010), Sakaida et

al., (2011), Takebayashi et al., (2013), Matsuo et al., (2016), and Hu et al.,

(2016).

The above-mentioned previous studies confirmed the mitigating effect of

sea breezes on areas of higher temperatures, and revealed that this influences

the UHI effect. Meanwhile, it has also been pointed out that in areas where

under normal circumstances higher temperatures would be mitigated by the

inflow of sea breeze, the mitigating effect is not exercised because clusters of

high-rise buildings block the inflow of sea breeze, and UHIs are formed.

However, the above-mentioned studies did not clarify in specific terms what

kind of effect clusters of high-rise buildings obstructing the inflow of sea

breezes has on the UHI effect.

Therefore, this study is unique in that it focuses on the obstruction of the

inflow of sea breezes by clusters of high-rise buildings and proposes a method

of investigating the influence of this on the UHI effect. Further, this study is

useful in that it proposes a very versatile method which can be applied to many

cities located on the coast. In the method of this study, using a weather

simulation model and a GIS, attention is focused on the relation of clusters of

high-rise buildings with the extent of rises in temperature, the positional

relationship of clusters of high-rise buildings with regions where the

temperature is higher and so on. Further, based on data on land use, landform

and climate, the influence which the obstruction of sea breeze by clusters of

high-rise buildings has on the UHI effect is quantitatively investigated.


3. FRAMEWORK AND METHOD

3.1. Framework

Firstly, the region and periods of time for investigation in this study is

selected in section 4, and data is collected and some data is processed in line

with the data format of the simulation model in section 5. Next, in section 6,

based on the various sets of data collected and processed in the previous

section, as the first stage of investigation, investigation of all the periods of

time subject to investigation is conducted, focusing on whether or not areas of

higher temperature caused by obstruction of sea breeze can be confirmed. In

section 7, as the second stage of investigation, based on the investigation

results from the first stage, attention is focused on periods of time in which sea

breeze can be confirmed. So that the situation regarding the obstruction of sea

breeze by clusters of high-rise buildings is better reflected, simulation is

conducted after part of the simulation model has been changed, and more

detailed investigation is carried out. Finally, in section 8, the findings of this

study are summarized, and the future research topics are mentioned.

3.2. Weather Simulation Model Selection

In this study, simulation was conducted focusing on a wide area;

therefore, the Weather Research and Forecasting (WRF) Model, which is a

mesoscale model, was selected. The WRF is a weather model whose

development is being advanced jointly by organizations such as the National

Center for Atmospheric Research (NCAR) and the National Centers for

Environmental Prediction (NCEP) in the United States (US). The WRF allows

simulation domain range and resolution to be set to suit the research aim, and

through the use of a canopy model, allows the influence of buildings on the

land surface to be analyzed in more detail. The WRF is starting to be widely

used in weather and climate research of recent years. Core versions of WRF

are the Nonhydrostatic Mesoscale Model (NMM), whose main purpose is

weather forecasting, and the Advanced Research WRF (ARW), whose main

purpose is academic research. In this study, based on Kusaka (2009, 2011), the

WRF-ARW Ver. 3.3.1 was used. Further, the WRF Preprocessing System

(WPS) is available as software for preparing data for executing the WRF, such

as initial conditions and boundary conditions. The WPS can be used to input

and edit data to set simulation domains. When working using WPS,


intermediate files are created at each stage of work, so by editing these files,

input data and so on can be changed and updated.

3.3. Method

In this study investigation was conducted in two stages, according to the

flow shown in Figure 1. Firstly, two scenarios which imagined different urban

forms - one with high-rise buildings and the other without high-rise buildings -

were created. Using Scenario 1, a simulation which imagined a fictitious urban

form in which there were no high-rise buildings at all and only low rise

buildings were distributed was conducted, and results were obtained for a

weather simulation in which the influence of high-rise buildings was not

reflected. Using Scenario 2, a simulation of a realistic urban form in which

there was a mixture of high-rise and low-rise buildings was conducted, and

results were obtained for a weather simulation in which the influence of high-

rise buildings was reflected. Next, the simulation results for the two scenarios

were comparatively analyzed.

The outline of the method is as follows. In both the first and second stages

of the investigation, a GIS was used to display simulation results for Scenarios

1 and 2 which were obtained using the WRF on digital maps, and the

simulation results were comparatively analyzed. Further, as the GIS, ESRI

Inc.’s ArcGIS ver.10.2 was used. Additionally, the Kelvin scale is used to

denote temperature in two stages of investigation in sections 6 and 7 (0°C =

273.15 K).

1) In the first stage, all the periods of time subject to investigation were

targeted. Simulations for all the time periods subject to investigation

were conducted for each of the two scenarios; focusing on

temperature and wind speed, the simulation results were

comparatively analyzed; and influence to the UHI effect was

investigated. Specifically, of the simulation results for each scenario,

temperature differences and wind speed differences were found, and

superimposed with high-rise building distribution data. This enabled

the influence of high-rise buildings on temperature and wind speed to

be investigated. Further, based on the comparative analysis results,

conditions for conducting more detailed investigation in the second

stage of investigation were clarified.


2) In the second stage of investigation, simulations were conducted for

each of the two scenarios, reflecting the conditions for conducting

detailed investigation which were found in the first stage of

investigation. Next, as in the first stage, these simulation results were

comparatively analyzed, and comparison was made with investigation

results from the first stage. This enabled the influence of high-rise

buildings on temperature and wind speed to be investigated. Based on

the investigation results, the obstruction of sea breeze by high-rise

buildings and the rises in temperature caused by this were clarified.

4. SELECTION OF REGION AND PERIODS OF TIME

FOR INVESTIGATION

4.1. Selection of Region for Investigation

As the region for investigation in this study, a region of 30 km square

centering on the Shiodome area of Minato City in the Tokyo Metropolis was

selected. Figure 2 shows the location of study region. In Figure 2, the square

of a gray frame shows the range of the region for investigation. Reasons for

selection were that in this region, there is a concentration of clusters of high-

rise buildings on the coastal areas of Tokyo Bay, and in guidelines published

by the Japanese Ministry of the Environment (2013), it is pointed out that

clusters of high-rise buildings are influencing the wind environment in this

region. As mentioned in section 1, the Shiodome area is a world-famous

example of an area where the forests of high-rise buildings in coastal areas

obstruct the flow of sea breezes into inland areas.

4.2. Outline of the UHI Effect on Human Health Problems

in Study Region

Mega-cities are defined by the United Nations (UN) as cities with more

than 10 million inhabitants. According to this definition, there were 20 mega-

cities worldwide in 2010. However, in 1975 there were just three mega-cities:

Tokyo, New York, and Mexico City. In Asia, many countries exhibit a

tendency for people to move from regional areas to large cities, mainly due to

high economic growth. As a result, there are presently 10 mega-cities in Asia:


two each in Japan and China, and three in India. The UN expects that the

Tokyo metropolitan population will increase to about 37 million by 2025.

Japan’s population is approximately 130 million people, and 51% of these live

in the three major metropolitan areas of Tokyo, Osaka-Kobe-Kyoto and

Nagoya. Since 28% of the Japanese population is concentrated in the Tokyo

metropolitan area, this phenomenon is referred to as “Tokyo Centralization.”

Japan has problems with a declining birthrate and aging population.

Nevertheless, the Japanese government predicts that the population will

continue to concentrate in the three major metropolitan areas, and that

depopulation of regional areas will progress in parallel. To address such

national-level problems, many Japanese cities, especially regional centers,

have recently adopted a compact city model, and the Japanese government

encourages emigration from metropolitan areas to regional areas.

Figure 1. Study flow.

As the Building Standard Law of Japan was relaxed after the 1960s, high-

rise buildings were built in quick succession. According to the Tokyo

statistical yearbook (2015), there are presently 535 high-rise buildings

measuring more than 100 m in the central Tokyo district (defined as the 23

wards of Tokyo). Especially in the coastal areas of Tokyo Bay, forests of high-


rise buildings measuring more than 200 m have been built and are dubbed the

“Tokyo Wall” (mentioned in section 1). Figure 3 shows representative areas

with dense concentrations of such forests of high-rise buildings. The locations

of these areas are shown in Figure 2. The Shiodome area, in particular, has

seen extensive redevelopment since 2002, resulting in construction of a huge

high-rise complex called Shiodome Sio-Site. Coastal areas adjacent to the

Shiodome area may experience redevelopment, which will further increase the

number of high-rise buildings in the near future. These forests of high-rise

buildings in coastal areas obstruct the flow of sea breezes into inland areas and

raise temperatures to unprecedented levels, intensifying the UHI phenomenon,

especially in the central area containing the 23 wards of Tokyo. This trend has

caused recorded temperatures to rise by approximately 3°C in the past 100

years. Given that this rise is high in comparison to other large cities in Japan

and other countries, urban warming due to the UHI effect is understood to be

remarkably high in addition to the effects of global warming.

Figure 2. Location of study region.

Figure 4 shows the trends of extremely hot days, tropical nights, and

deaths caused by heatstroke from 1980 to 2015 in Japan. The Japan

Meteorological Agency defines a tropical day as a day on which the

temperature goes above 35°C, and a tropical night as a night on which the

minimum temperature remains above 25°C. As can be seen from Figure 4, the


numbers of extremely hot days, tropical nights, and deaths caused by

heatstroke in 2010 were the highest for the past 35 years. According to the

Japanese Ministry of Health, Labour and Welfare (2016), the numbers of

deaths caused by heatstroke in the Tokyo metropolitan area were 179 in 2013

and 73 in 2014, and the UHI effect is causing increasing harm to elderly

people in particular. The number of deaths caused by heatstroke in the 23

wards of Tokyo has increased tremendously compared to other Japanese cities.

Shiodome area

Odaiba area

Shibaura area

Figure 3. (Continued).


Shinagawa area

Figure 3. Representative areas with dense concentrations of high-rise building

“forests” in study region.

Furthermore, Kusaka et al., (2013) used the dynamical downscaling

method with the WRF model to predict the future summer climate in the three

Japanese metropolises of Tokyo, Nagoya, and Osaka under the Special Report

on Emission Scenarios (SRES) A1b scenario assumed by the

Intergovernmental Panel on Climate Change (IPCC). Impacts of climate

change on human health were examined by a mid-point impact assessment

based on the contingent valuation method. The results show that the frequency

of days with daytime maximum temperatures exceeding 35°C will double by

the 2070s in these three metropolises. Accordingly, there will be an increase in

people suffering from health impediments such as heatstroke, especially in

these three metropolises.

4.3. Selection of Periods of Time for Investigation

Concerning the periods of time for investigation, taking the aim of this

study into consideration, August 7th from the record-breaking heat wave of

July to August 2010 was selected as a date that satisfied all the

undermentioned conditions of fine weather days and sea breeze days defined

by Kitao et al., (2012). As mentioned in section 4.2, in Japan, the numbers of

extremely hot days, tropical nights, and deaths caused by heatstroke in 2010

were the highest for the past 35 years.

1) Fine weather days:

The weather is fine or sunny (The average degree of cloudiness is

8.4 or less)


here are seven or more hours of sunlight

The daily amount of accumulated global solar radiation is 19

MJ/m2 or more

The amount of precipitation is less than 0.5 mm

2) Sea breeze days:

In the 12 hours between midday and midnight, the wind speed is

2 m/s or more and there is a sea breeze for six hours or more.

Further, the weather data used in the simulations which employ the WRF

are data for six hour intervals for 3 AM, 9 AM, 3 PM and 9 PM of the Japan

Standard Time. Because it is necessary to make these times the start and finish

times of the simulations, the specific period of time for which simulation was

conducted was 3 AM on August 7th, 2010 to 3 AM on August 8th, 2010 (Japan

Standard Time).

5. DATA COLLECTION AND PROCESSING

5.1. Data Collection

In weather simulations which use the WRF, a group of open big data in

environmental science such as land cover data constituted from land use data

and landform data and weather data, atmospheric data, sea surface temperature

is necessary. In the WRF, as default data, for land cover data, data from the

US Geological Survey (USGS) is input; and as weather data, objective

analysis data from the NCEP (Global Final Analyses: FNL) is input. Besides

these sets of default data, in Japan, land use data from the Geospatial

Information Authority of Japan (GSI) can be used as land use data, and meso

objective analysis data from the Japan Meteorological Agency (Meso-Scale

Analysis: MANAL) can be used as weather data. Akimoto et al., (2010)

showed that by using land use data from the GSI instead of data from the

USGS, calculation results that more closely match observed values can be

obtained in Japan. Therefore, in this study, as weather data and landform data,

the WRF default data, that is, FNL and USGS data, were used; and as land use

data, based on Akimoto et al., (2010), urban area land use subdivided mesh

data for 2009 from the GSI was used.

Note: Based on surveys by the Japan Meteorological Agency (2016) and the Japanese Ministry of Health, Labour and Welfare (2016).

The 2015 figure for number of deaths caused by heatstroke is as of late August.

Figure 4. Trends of extremely hot days, tropical nights and deaths caused by heatstroke in Japan (1980-2015).

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

0

10

20

30

40

50

60

198

0

198

1

198

2

198

3

198

4

198

5

198

6

198

7

198

8

198

9

199

0

199

1

199

2

199

3

199

4

199

5

199

6

199

7

199

8

199

9

200

0

200

1

200

2

200

3

200

4

200

5

200

6

200

7

200

8

200

9

201

0

201

1

201

2

201

3

201

4

201

5

Extremely hot day Tropical night Deaths caused by heatstroke

Days Deaths

Figure 5. Urban forms in Scenarios 1 and 2 in study region.


Table 1. Correspondence List of Land Use Classification

Land use classification for

WRF

Urban area land use subdivided mesh data

Scenario 1 Scenario 2

Urban and built-up land Road Road

Railway Railway

Dryland, cropland and pasture - -

Irrigated cropland and pasture Rice field Rice field

Mixed dryland/Irrigated

cropland and pasture

- -

Cropland/Grassland mosaic Wasteland Wasteland

Cropland/Woodland mosaic - -

Grassland Other farmlands Other farmlands

Golf course Golf course

Open space Open space

Park and greenery area Park and greenery area

Shrubland - -

Mixed shrubland/Grassland - -

Savanna - -

Deciduous broadleaf forest - -

Deciduous needleleaf forest - -

Evergreen broadleaf forest - -

Evergreen needleleaf forest - -

Mixed forest Forest Forest

Water bodies River and lake River and lake

Seashore Seashore

Sea Sea

Herbaceous wetland - -

Wooded wetland - -

Barren of sparsely vegetated - -

Herbaceous tundra - -

Wooded tundra - -

Mixed tundra - -

Bare ground tundra - -

Snow or ice - -

Low intensity residential Low intensity building Low intensity building

High intensity building -

High intensity residential - High intensity building

Industrial of commercial Factory Factory

Public facilities Public facilities


5.2. Data Processing

Urban area land use subdivided mesh data from the GSI was processed

according to the following procedure:

(1) There are 27 categories of land use in the WRF; however, the above-

mentioned urban area land use subdivided mesh data has 11 categories, and its

method of classifying land use is different to that of the WRF. Therefore, with

reference to Akimoto et al., (2010), the land use categories in the urban area

land use subdivided mesh were made to correspond to the unique categories of

the WRF, as shown in Table 1. Further, in Scenario 1, which does not divide

buildings into low-rise buildings and high-rise buildings, both types of

buildings were classified as “Low intensity building.” In Scenario 2, which

divides buildings into low-rise buildings and high-rise buildings, low-rise

buildings were classified as “Low intensity building,” and high-rise buildings

were classified as “High intensity building.” This work was performed using

the ArcGIS field calculation function, and the processed land use data was

output in CSV file form. Urban forms in Scenarios 1 and 2 in the region for

investigation are shown in Figure 5.

Further, when working with the WRF, it is necessary to create variables

which show whether surfaces in the simulation domain are water surfaces or

land surfaces. Therefore, ArcGIS was used to create data in which land uses

which correspond to “Water bodies” in Table 1 were classified as water

surfaces, and land uses other than these were classified as land surfaces. This

data was also output in CVS file form.

(2) Using the WPS, values for the variable LU_INDEX (which shows land

use) and values for the variable LANDMASK (which shows the state of the

surface) in the intermediate files of the simulation domains created using the

USGS land cover data were substituted with values from the CSV files output

in step (1) described above, and simulation domains which reproduced the

land use of each scenario were created. Similarly, using the WPS, the weather

data was also input into the simulation domains, and files for input into the

WRF were created.

(3) Using an application for the creation of WRF boundary files, vertical

boundary data was input into the files for input into the WRF that were created

in step (2), described above. At this time, the variable IVGTYP, which shows

vegetation of land surface, was newly added. Values for this variable should

have been the same as those for LU_INDEX; however, when the data created

for input into the WRF was checked, it was found that some of the

LANDMASK values had changed into values different to those used for


substitution in step (2) above. Therefore, the values which differed were

corrected by being once again replaced with the CSV file values output in step

(1).

6. INVESTIGATION OF ALL THE PERIODS OF TIME IN

THE FIRST STAGE

6.1. Post-Processing of Simulation Results

Simulation results for wind direction and wind speed were illustrated

using the GIS. Using the WRF, wind direction and wind speed can be

calculated as scalar quantities for each simulation domain grid cell, based on

the east-west direction (x component) wind speed and the north-south

direction (y component) wind speed, as shown below.

1) Calculation of wind speed

Wind speed is expressed by the magnitude of a vector formed from an x

component and a y component. Therefore, it can be found using the following

equation:

𝑊𝑖𝑛𝑑 𝑠𝑝𝑒𝑒𝑑 = √𝑥2 + 𝑦2 (1)

2) Calculation of wind direction

Wind direction can be found as a gradient θ of a vector in a mathematical

x-y plane, as shown in Figure 6. The gradient can be found using the inverse

trigonometric function arctan; however, the range of θ calculated using arctan

is π/2 ≧ θ ≧ -π/2, and not all the gradients of the range can be obtained.

Therefore, using the atan2 function of the mathematical problem solver library

of the programming language Python, which is included in the ArcGIS

ver.10.2, the gradient was found using the following equation:

θ = atan2(y⁄x) (2)

Sasaki et al., (2005), Kitao et al., (2012), and Grawe et al., (2012)

conducted simulation using a grid with grid cell intervals of several


kilometers. Similarly to these previous studies, in this study, based on

simulation results output for each 100 m interval grid cell, average values for

each 1 km interval grid cell were found and resampled in order to derive the

wind direction.

3) Calculation of temperature differences and wind speed differences

between Scenarios 1 and 2

Further, in order to compare the simulation results of Scenarios 1 and 2,

differences in temperature and wind speed between Scenarios 1 and 2 were

also found for each 1 km interval grid cell. As in studies by Taniguchi et al.,

(2008) and Junimura et al., (2008), the difference in wind speed was derived as

a scalar quantity.

Figure 6. Calculation of wind direction.

6.2. Simulation Results for Each Hour

In the investigation for all the periods of time subject to investigation,

simulation results were organized for each hour from 4 AM to midnight.

Figure 7 shows wind direction and wind speed for Scenario 2 (the scenario

with a realistic urban form in which there is a mixture of high-rise buildings

and low-rise buildings) for 6 AM to 9 AM. From 4 AM to 6 AM the wind was

not a sea breeze; however, as shown in Figure 7, after a calm period from 7

AM to 8 AM, the wind direction changed and the wind became a sea breeze,


and it was possible to confirm a sea breeze from about 9 AM to 9 PM. Further,

Figure 8 shows temperature differences and wind speed differences between

Scenarios 1 and 2 from 7 PM to 9 PM. In Figure 8, temperature differences

and wind speed differences between Scenarios 1 and 2 from 7 PM to 9 PM are

each superimposed with the distribution of high-rise buildings. As will be

described in detail below, in this time period it was possible to confirm a rise

in temperature and a reduction in wind speed in a region downwind of high-

rise buildings.

6.3. Discussion

Simulation results for Scenario 1 (which had an imaginary urban form

which did not include high-rise buildings), and Scenario 2 (which had a

realistic urban form which included a mixture of high-rise buildings and low-

rise buildings) were comparatively analyzed, focusing on temperature

difference and wind speed difference. In the analysis, the influence of high-

rise buildings on temperature and wind speed was investigated by showing

how temperature and wind speed increased and decreased in Scenario 2 in

relation to Scenario 1, based on differences in temperature and wind speed

between Scenarios 1 and 2, as illustrated in Figure 8.

From 9 AM to 6 PM, both increases and decreases in wind speed caused

by high-rise buildings were detected. Regions where wind speed dropped and

rose were located in no particular pattern in the inland area. Further, regions

where increases in temperature occurred were sparsely located in the inland

area, and the degree of the increase in temperature was from 0.2 to 1 K.

However, because regions where wind speed dropped and regions where

warming occurred were not necessarily the same, it is difficult to say that

increases in temperature were caused by the obstruction of sea breeze by high-

rise buildings. However, as shown in Figure 8, an increase in temperature of

about 0.2 K was confirmed in the time period 7 PM to 9 PM in a region of 5 to

10 km square downwind of high-rise buildings. Meanwhile, there were regions

where the wind speed dropped in a region 2 km square downwind of high-rise

buildings, but compared to regions of higher temperature, their range was very

small. Therefore, it can be said that in a region 2 km square downwind of

high-rise buildings where a rise in temperature and a reduction in wind speed

were confirmed, a higher (by about 0.2 K) temperature occurred due to the

obstruction of sea breeze. Mikami (2005) pointed out that the rise in

temperature caused by urbanization alone in the Tokyo Metropolis has been


about 2 K over the past hundred years. From the above-mentioned simulation

results of this study, it is clear that the rise in temperature due to obstruction of

sea breeze by high-rise buildings is about 0.2 K, which is equivalent to about

10% of the rise in temperature caused by urbanization.

Changes in wind speed that were due to the presence or absence of high-

rise buildings were confirmed in many of the time periods between 4 AM and

midnight. However, wind speed differences in the period 10 AM to 5 PM were

confirmed regardless of the presence or absence of high-rise buildings.

Further, from 6 PM onwards, there is a region where a decrease in wind speed

occurred downwind of high rise buildings, so it was possible to confirm

obstruction of sea breeze by high-rise buildings; however, the range of this

area is smaller than that of the region where there was a rise in temperature.

Therefore, it is considered that the influence of high-rise buildings on wind

speed was not sufficiently reproduced by the simulation. Considering the

simulation results in more detail, the height of high-rise buildings in the urban

canopy model introduced into the WRF is originally set at 7.5 m, whereas the

calculated values for wind speed output by the WRF are values for 10 m above

the ground. Therefore, due to the original setting of the WRF, it is clear that

the influence of high-rise buildings on wind speed was not sufficiently shown

in the simulation results in this section. Accordingly, it was thought that if the

urban canopy model parameters were changed such that the height of high-rise

buildings was 10 m or more, obstruction of sea breeze would show up more

clearly in the simulation results. Therefore, in the next section, high-rise

building height is set at 10 m, and the time period 7 PM to 9 PM, in which sea

breeze was confirmed in this section, is investigated in detail.

7. INVESTIGATION OF TIME PERIODS IN WHICH SEA

BREEZE OCCURRED IN THE SECOND STAGE

7.1. Simulation Results

Based on the results in the previous section, all conditions were made the

same as in the previous section, except for high-rise building height, which

was changed to 10 m, and simulation was conducted for the period 9 AM to 9

PM. Figure 9 shows wind speed differences between Scenarios 1 and 2 for 6

PM in the first stage of the investigation (described in the previous section)

and the second stage of the investigation (described in this section). From


Figure 9, it can be seen that setting the high-rise building height to 10 m

resulted in the influence of high-rise buildings on wind showing more clearly

in the simulation results of this section than it did in the simulation results of

the previous section. Further, Figure 10 shows temperature differences and

wind speed differences between Scenarios 1 and 2 from 6 M to 9 PM. In

Figure 10, temperature differences and wind speed differences between

Scenarios 1 and 2 for the time period 6 PM to 9 PM are each superimposed

with high-rise building distribution. As will be described in detail below, it

was possible to confirm higher temperatures caused by obstruction of sea

breeze by high-rise buildings in this time period.

7.2. Discussion

For the time period 9 AM to 9 PM, as in the previous section, simulation

results for Scenarios 1 and 2 were comparatively analyzed, and comparison

with investigation results from the previous section was conducted. In the

comparative analysis of simulation results for Scenarios 1 and 2, as in section

6.3, the influence of high-rise buildings on temperature and wind speed was

investigated by showing how temperature and wind speed increased and

decreased in Scenario 2 in relation to Scenario 1, based on differences in

temperature and wind speed between Scenarios 1 and 2, as illustrated in Figure

10.

Firstly, in the time period 1 PM to 4 PM, there is a region containing a

mixture of increases and decreases in wind speed; however, it was confirmed

that the wind speed decreases in approximately the same region as the region

where high-rise buildings are located. Further, in the time period 5 PM to 9

PM, concerning a decrease in wind speed that was barely identifiable in the

simulation results of the previous section, as shown by the results for 7 PM to

9 PM in Figure 10, in the simulation results of this section it became possible

to confirm decreases in wind speed in a region of high-rise buildings and a

region downwind of high-rise buildings. These results show that setting the

high-rise building height to 10 m allowed the influence of high-rise buildings

on wind speed to be reflected in the simulation results, and it was revealed that

sea breeze was obstructed in a region of high-rise buildings and a region

downwind of high-rise buildings.

Figure 7. Wind direction and wind speed for Scenario 2 for 6 AM to 9 AM.

Figure 8. Temperature differences (left side) and wind speed differences (right side) between Scenarios 1 and 2 from 7 PM to 9 PM.

Figure 9. Wind speed differences between Scenarios 1 and 2 for 6 PM in the first stage of the investigation.

Figure 10. Temperature differences (left side) and wind speed differences (right side) between Scenarios 1 and 2 from 6 M

to 9 PM.


When the height of high-rise buildings was set to 10 m in the investigation

in this section, the result was a decrease in the range of the region containing a

mixture of increases and decreases in wind speed in the time period 9 AM to 4

PM; however, the region still existed, as in the investigation of the previous

section. The existence of turbulence is a possible reason for regions with a

mixture of increases and decreases in wind speed occurring in this manner.

Based on Takeuchi (1997) and Kondou (2000), it is thought that vertical winds

occur due to high-rise buildings intensifying turbulence; that horizontal wind

speed decreases in regions where the wind flows upwards and increases in

regions where the wind flows downwards; and that in regions with high-rise

buildings, there is a mixture of these two types of region.

At 5 PM in a region with a wind speed lower by about 2 m/s, it was

confirmed the temperature was higher by about 0.3 K. From 6 PM to 9 PM the

decrease in wind speed was about 1 m/s; however, the range of the area where

the wind speed decreased spread further. In conjunction with this, the

temperature was higher by about 0.2 K over a wide range. Since a decrease in

wind speed and a higher temperature were confirmed at the same time over a

wide range in this manner, it is clear there is a relationship between decrease

in wind speed and higher temperatures. Therefore, it is thought that in the

region where it was possible to confirm a decrease in wind speed and a higher

temperature at the same time during 5 PM to 9 PM, a higher temperature

caused by obstruction of sea breeze by high-rise buildings occurred, and it

occurred over an area of 5 to 10 km square.

As described above, in the period of time subject to investigation in this

section, it was possible to confirm a higher temperature caused by obstruction

of sea breeze during the period 5 PM to 9 PM. In the period 9 AM to 4 PM,

decreases and increases in wind speed thought to have been caused by

turbulence were confirmed, and in these areas there were also areas of higher

temperature; therefore, it is difficult to say that the higher temperatures

occurred due to obstruction of sea breeze by high-rise buildings in the period 9

AM to 4 PM. During this period of time, the temperature on land was high, at

304.6 K or more, and wind speed was on the rise. However, in the period 5

PM to 9 PM, the temperature continued to decrease from 5 PM onwards, and

wind speed continued to decrease from 6 PM onwards. After 5 PM to 6 PM,

during which warming caused by obstruction of sea breeze by buildings can be

confirmed, temperature and wind speed in the study region (the central Tokyo

district) were on the decline. As the temperature decreased from 5 PM

onwards, it became possible to confirm higher temperatures caused by

obstruction of sea breeze; therefore, it is thought that although normally, the


temperature on land should cool down due to the inflow of sea breeze, an area

of higher temperature occurred due to sea breeze being blocked in a region

downwind of high-rise buildings.

Therefore, it is clear that rather than impeding the suppression of increases

in daytime temperature, the obstruction of sea breeze by high-rise buildings

weakens the decrease in temperature which occurs from evening onwards, and

contributes to the formation of nighttime UHIs. Furthermore, it is also pointed

out that the nighttime UHIs causes a tropical night which may affect human

health problems during the periods of time for investigation. Incidentally, as

mentioned in section 4.2, in Japan, the numbers of extremely hot days, tropical

nights, and deaths caused by heatstroke in 2010 were the highest for the past

35 years.

CONCLUSION AND FUTURE RESEARCH TOPICS

Focusing on the UHI which has caused human health problems and the

issue that sea breeze flows are being obstructed by forests of high-rise

buildings in coastal areas, this study aims to investigate the influence on the

UHI effect of the obstruction of sea breezes by clusters of high-rise buildings

in the central Tokyo district of Japan. To that end, using a weather simulation

model in addition to GIS, this study quantitatively analyzed a group of open

big data in environmental science such as land cover data constituted from

land use data and landform data and weather data, atmospheric data, sea

surface temperature.

The findings of this study are summarized in the following three points.

1) A method of investigating the influence of the obstruction of sea

breeze by high-rise buildings on the UHI effect was proposed. In the

method, two scenarios that imagine urban forms which differ with

regard to whether or not they contain high-rise buildings are created

and weather simulation is conducted, and the results of the

simulations are comparatively analyzed focusing on temperature and

wind speed. Investigation was conducted in two stages, and a region

of 30 km square centering on the Shiodome area of Minato City in the

Tokyo Metropolis and a date from the record-breaking heat wave of

July to August 2010 were selected for investigation. In Japan, the

numbers of extremely hot days, tropical nights, and deaths caused by

heatstroke in 2010 were the highest for the past 35 years. The number


of deaths caused by heatstroke in the 23 wards of Tokyo has increased

tremendously compared to other Japanese cities.

2) In the first stage, the influence of high-rise buildings on temperature

and wind speed was investigated in all the periods of time, and

conditions for conducting more detailed investigation were found. The

first stage of investigation revealed that during the period of time

subject to investigation, sea breeze could be confirmed from 9 AM to

9 PM. Further, a rise in temperature of approximately 0.2 K caused by

the obstruction of sea breeze was confirmed in an area 2 km square

downwind of high-rise buildings. In addition, the first stage of

investigation revealed the necessity of setting the height of high-rise

buildings in the weather simulation model to suit present conditions of

urban regions in Japan, and conducting detailed investigation in the

second stage of investigation.

3) In the second stage, based on the conditions clarified in the first stage,

the influence of high-rise buildings on temperature and wind speed

was investigated in detail, and the obstruction of sea breeze by high-

rise buildings and the higher temperature that is caused by this were

revealed. In the second stage of investigation, a rise in temperature of

approximately 0.3 K and a reduction in wind speed of approximately

1 m/s were observed in a region of 5 to 10 km square downwind of

high-rise buildings in the period 6 PM to 9 PM, and a higher

temperature caused by the obstruction of sea breeze by high-rise

buildings was identified. The fact that such a higher temperature was

confirmed in the time period from 6 PM onwards, in which the

temperature decreases, reveals that obstruction of sea breeze by high-

rise buildings dulls the decrease in temperature which occurs from

evening onwards, and influences nighttime UHI formation. Therefore,

the UHI causes a tropical night which may affect human health

problems.

Examples of future research themes are to apply the method proposed in

this study to other regions and periods of time and verify its validity, and to

check detailed changes in temperature and wind speed by conducting weather

simulation using shorter intervals of time.


REFERENCES

Akimoto, Y. and Kusaka, H. (2010). Sensitivity of the WRF regional

meteorological model to input datasets and surface parameters for the

Kanto Plain on fine summer days, Geographical Review of Japan, 83(3),

324-340.

Grawe, D., Thompson, H. L., Salmond, J. A., Cai, X. M. and Schlünzen, K. H.

(2012). Modelling the impact of urbanisation on regional climate in the

Greater London Area, International Journal of Climatology, 33(10), 2388-

2401.

Hu, X. M. and Xue, M. (2016). Influence of synoptic sea-breeze fronts on the

urban heat islandintensity in Dallas - Fort Worth, Texas, Monthly Weather

Review, 144, 1487-1507.

Ihara, T., Kusaka, H., Hara, M., Matsuhashi, R. and Yoshida, Y. (2011).

Estimation of mild health disorder caused by urban air temperature

increase with midpoin-type impact assessment methodology, Journal of

Environmental Engineering (Transactions of AIJ), 76(662), 459-467.

Imai, K. and Yamamoto, K. (2010). An evaluation of measures against the

urban heat island from the viewpoint of artificial exhaust heat of road

traffic - An evaluation using GIS in the Tokyo 23-Ward Area-, Japan

Society of Socio-Informatics journal, 21(2), 47-58.

Imai, K. and Yamamoto, K. (2015). An evaluation of measures regarding road

traffic against the urban heat island in the Tokyo Ward Area, Japan,

Journal of Environmental Protection, 6(9), 957-975.

Japan Meteorological Agency. (2016). Heat Island Monitoring Report (2015),

Tokyo.

Japanese Ministry of the Environment. (2013). Urban heat island

countermeasures guideline for the year 2012-revised edition, http://www.

env.go.jp/air/life/heat_island/guideline/h24.html, Accessed 22 January,

2014.

Japanese Ministry of Health, Labour and Welfare. (2016). Number of deaths

caused by heatstroke, http://www.mhlw.go.jp/stf/seisakunitsuite/bunya/00

00121413.html, Accessed 17 August, 2016.

Junimura, Y. and Watanabe, H. (2008). Study on the mitigating effects of sea

breeze on urban temperatures in summer - Analyses based on long-term

multi-point simultaneous observation of temperature and observed wind

data-, Journal of Environmental Engineering (Transactions of AIJ), 73

(623), 93-99.


Kato, A., Okitsu, Y., Tsunematsu, N., Hongyo, T., Kobayashi, T. and

Ichihashi, A. (2015). Mitigation effect of urban heat island from forest

canopy structure of urban forest, Journal of the Japanese Society of

Revegetation Technology, 41(1), 169-174.

Kitao, N., Moriyama, M., Takebayashi, H. and Tanaka, T. (2012). Study on

method for creating an urban environment climate map for the Osaka

region, AIJ Journal of Technology and Design, 18(38), 255-258.

Kondou, J. (2000). Atmospheric Science near the Ground Surface, University

of Tokyo Press: Tokyo.

Kusaka, H. (2009). The regional atmospheric model WRF, Journal of Japan

Society of Fluid Mechanics, 28, 3-12.

Kusaka, H. (2011). Application of the regional atmospheric model WRF to

urban climate study and future issues, Journal of Geography, 120(2), 285-

295.

Kusaka, H., Iijima, N., Ihara, T., Hara, M., Takane, Y. and Iizuka, S. (2013).

Future projection of heat stroke and sleep distribution for 2070’s august in

Tokyo, Nagoya, and osaka - Dynamical downscale experiments from

multiple CMIP3-GCMs and health impact assessment by mid-point type

methodology-. Journal of Environmental Engineering (Transactions of

AIJ), 78(693), 873-881.

Matsuo, K. and Tanaka, T. (2013). Analysis on summer daytime temperature

distribution in coastal city where sea breeze affects: Analyses based on

long-term multi-point measurements in Hiroshima, Journal of the City

Planning Institute of Japan, 48, 819-824.

Matsuo, K. and Tanaka, T. (2016). Aanaysis on the effect of sea breeze on

summer diurnal temperature distribution pattern in Hiroshima Plain -

Mapping the sea breeze effects for mitigating urban warming-, Journal of


Mikami, T. (2005). Urban heat island phenomena and their causing factors - A

case study of the Greater Tokyo Metropolitan Area-, Journal of

Geography, 114(3), 496-506.

Nakagawa, K. (2011). Trends in studies on the formative factors of the urban

heat island in Japan, with special emphasis on the formative factors of

urban heat island intensity, Journal of Geography, 120(2), 255-284.

Nobayashi, A. and Hayashi, Y. (2011). Observational study on effect of

dynamical mixing on small-scale heat islands, Bulletin of the Terrestrial

Environment Research Center, the University of Tsukuba, 10, 75-81.


Omura, T. and Nakano, T. (2009). Report of analyzing the relationship

between heat illness and the heat island phenomenon, Environmental

Information Science, 40(2), 70-74.

Ooka, R., Satou, T. and Murakami, S. (2008). Numerical simulation of sea

breeze over the Kanto Plain and analysis of the interruptive facters for the

sea breeze based on mean kinetic energy balance. Journal of


Papanastasiou, D. K., Melas, D. and Bartzanas, T. (2010). Temperature,

comfort and pollution levels during heat waves and the role of sea breeze,

International Journal of Biometeorology, 54(3), 307-317.

Sakaida, K., Egoshi, A. and Kuramochi, M. (2011). Urban heat islands and the

effects of sea breezes in Sendai, Journal of Geography, 120(2), 382-391.

Susca, T., Gaffinb, S. R. and Dell’Ossoa, G. R. (2010). Positive effects of

vegetation: Urban heat island and green roofs, Environmental Pollution,

2119-2126.

Takahashi, K. and Takahashi, H. (2013). Influence of urban heat island

phenomenon in the central tokyo on nocturnal local wind system in

summer: A case study using atmospheric pressure data of high density

observation network, Tenki, 60(7), 505-519.

Takebayashi, H. and Moriyama, M. (2013). Study on the effects of

urbanization on the cooling effects of the sea breeze, Journal of


Takeuchi, K. (1997). Meteorology of the Wind, University of Tokyo Press:

Tokyo.

Tan, J., Zheng, Y., Tang, X., Guo, C., Li, L. and Song, G. (2010). The urban

heat island and its impact on heat waves and human health in Shanghai,

International Journal of Biometeorology, 54(1), 75-84.

Taniguchi, A., Kiyota, N. and Kiyota, T. (2008). Study on mitigation effect of

sea wind for high temperature phenomenon in urban area in coastal city,

Journal of Environmental Engineering (Transactions of AIJ), 73(625),

379-384.

Tokyo Metropolitan Government. (2015). Tokyo statistical yearbook 2015,

http://www.toukei.metro.tokyo.jp/tnenkan/tn-eindex.htm, Accessed 17

August, 2016.

Troy, F., Sadohara, S. and Yoshida, S. (2009). Designing landuse patterns to

mitigate the urban heat island phenomenon in consideration of ocean wind

flow and openspace voids, Journal of Architecture and Planning, 74(644),

2223-2229.


Tyson, P. D. and Preston-Whyte, R. A. (1973). Temperature structure of city

sky: Paper review and survey plan of urban heat island effect in

Johannesburg, Journal of Architecture and Building Science, 1062, 258-

259.

Uenoyama, A. and Murahashi, M. (2013). Ventilation path by using three

dimensional simulation - Case study of Osaka’s urban areas-. Journal of

the City Planning Institute of Japan, 48(3), 921-926.

Wong, M. S., Nichol, J. and Ng, E. (2011). A study of the “wall effect” caused

by proliferation of high-rise buildings using GIS techniques, Landscape

and Urban Planning, 102, 245-253.

Yamato, H., Mikami,T. and Takahashi, H. (2011). Influence of sea breeze on

the daytime urban heat island effect in summer in the Greater Tokyo

Metropolitan Area, Journal of Geography, 120(2), 325-340.

Yokoyama, M., Matsuo, K., Tanaka, T. and Sadohara, S. (2014). A study on

the urban climate under the compact city scenario by using meso-scale

meteorological model: Scenario evaluation in Kanagawa Prefecture,

Journal of the City Planning Institute of Japan, 49(3), 915-920.

Kayoko Yamamoto1,*, PhD, and Shota Sashiyama2...Tokyo is the largest mega-city in the world...

Documents

Transcript of Kayoko Yamamoto1,*, PhD, and Shota Sashiyama2...Tokyo is the largest mega-city in the world...